two schools of thought on the nsa. brian, interesting the obama administration pushed back on "the washington post," saying don't run that story, at least let us talk about what you're breaking. i find that fascinating. >> so many moving parts to this. here's the thing. i am fascinated by the fact peter king and michael hayden have to further define the program, as soon as pushed by "the washington post" or telegraph, it is unbelievable. they point out as michael hayden did, if you look at the amount of work the nsa is doing with the percentage of mistakes and they all agree, it is human error, it is not some problem with the system, he's saying it is 99.9% accurate. what i have a problem is not being transparent to the fisa court. >> last year, 3,000 mistakes? >> out of 240 billion? >> no, no, no, no, no. that's one instance of one office, of one server they picked up 3,000 errors. kimberly, bob, do you want to weigh in on this? >> i am scared of you today. >> me, too. >> yes, by the way, well -- >> don't. >> listen,